Project goal:

The objective of this Tableau project was to create a scenario where key data insights can be easily viewed by Airbnb business stakeholders that center on various Seattle Residential zip codes in 2016.

Data source:

Source of data set was found from this hyperlink:

2016 Airbnb Housing Data

Business insights from data set:

Airbnb stakeholders primarily requested home analysis of their 2016 raw data set due to their findings of missing zip code values that were present within the Listings worksheet.

• This request of Airbnb stakeholders was answered due to the data cleaning techniques that were applied upon the zip codes data that was found in this project.

Some possible questions that Airbnb stakeholders may think of based on the information given in the dashboard:

• Which zip code in Seattle has the highest average home price?
  • Business insight:
    • The zip code, 98134, has the highest average home price in Seattle.
• Which bedroom count has the greatest number of homes available for consumers?
  • Business insight:
    • The highest number of homes available for consumers is, 1,811, when the bedroom count is one.
• Which months in 2016 had the highest revenue when compared to the other months?
  • Business insight:
    • The following months in 2016 had the highest revenue: March, May, June, and December.
      • Week of March 27th yielded a revenue profit of $1,906,735
      • Week of May 29th yielded a revenue profit of $2,013,698
      • Week of June 19th yielded a revenue profit of $2,073,319
      • Week of December 25th yielded a revenue profit of $2,110,350
        • This data collection suggests that high revenue levels are recorded when customers wish to check out Airbnb homes during the spring season and holiday seasons.

Import data set and data tools used:

Imported the data set into Microsoft Excel for data analysis and utilized Tableau for data visualization purposes.

Data set details:

Raw data set contained 323,346 entries.

The following three worksheets were found in the imported Microsoft Excel csv file:

1. Reviews
2. Listings
3. Calendar

Concentrated on the following columns per worksheet for data exploration purposes:

• Found within the Listings worksheet:
  • host_location
  • id
  • neighbourhood_cleansed
  • neighbourhood_group_cleansed
  • city
  • state
  • zipcode
  • latitude
  • longitude
  • bedrooms
  • price
  • weekly_price
• Found within the Reviews worksheet:
  • listing_id
• Found within the Calendar worksheet:
  • listing_id
  • date
  • price

Special mention: Recognizing patterns is pivotal to being successful in data analysis.

For example, in the Listings worksheet, there is column for id.

Within the Reviews worksheet, there is a column for listing_id

Finally, in the Calendar worksheet, a column for listing_id is present.

Overall, joining data sets for this project is necessary for accurate data visualization purposes.

Data Cleaning methods:

Overview of data cleaning methods that were utilized:

Created one new worksheet called:

• Working Sheet - Listings
  • Note: Editing raw data (Listings) worksheet is strongly not recommended.
  • Data industry best practice is to backup and keep raw data unedited

Data cleaning methods that were performed within Working Sheet - Listings

• Applied Microsoft Excel data filter button
  • Filtered for null values within the zipcode column
  • Discovered seven missing values for the following entries and replaced with the appropriate zip codes:

1. Entry 164

   • Value = 98107

2. Entry 481

   • Value = 98119

3. Entry 990

   • Value = 98122

4. Entry 1674

   • Value = 98104

5. Entry 1919

   • Value = 98119

6. Entry 2523

   • Value = 98102

7. Entry 2673

   • Value = 98102

Note:
I found the seven missing zip code values by cross referencing the following columns in an internet search:

neighbourhood_cleansed, neighbourhood_group_cleansed, city, state

Example of internet search that was performed:

• zip code for broadway seattle Washington
  • Repeatedly searched the internet for residential information in Seattle using different search parameters each time per the following listed below:
    • West Woodland
    • North Queen Anne
    • Yeslter Terrace
    • Interbay
    • Portage Bay

Tableau Data Preparation:

With data cleaning being completed, the next step was to move onto Tableau Public program where I would create data visualizations for Airbnb business insights purposes.

Initialized the Tableau Public program on my computer. Afterwards, I clicked on the option to open a Microsoft Excel file.

As a result, three tabs were presented to me:

1. Calendar
2. Working Sheet - Listings
3. Reviews

Before I start, as you may recall from earlier in the data set details section, I made a special note to mention recognizing patterns in data is essential to successfully join data sets so that there is accuracy in data visualizations.

The visual below illustrates the eight steps that were taken in the Tableau Public program to combine the data sets.

The following eight steps were taken to join the tables in Tableau Public program:

1. Click and drag the Working Sheet - Listings tab into the joining tables area of the program.
2. Click on the update automatically button to instruct the Tableau Public program to update Working Sheet - Listings table.
3. Clicked on the Working Sheet - Listings tab and opened up the tab to allow another tab field to join.
4. Click and dragged the Calendar tab to position it next to the Working Sheet - Listings tab.
5. An object containing two overlapping circles was placed between the two tabs, Working Sheet - Listings tab and Calendar; in addition, the two overlapping circles object was clicked on to continue onwards.
6. Inner join option was preselected which is acceptable because this will be the preferred method for data visualization purposes.
7. Selected listing_id under Calendar area and then selected price under data source area. After doing this step, the two tables were joined to form 96 fields with 1,048,575 rows.
8. Final step, I clicked on Sheet 1 at the very bottom of the Tableau Public program; in which, this instructs the Tableau Public program that the user is ready to start creating data visualizations.

Creation of Tableau Dashboard elements:

Overview of designing Tableau dashboard elements:

Sheet 1

• Title: Advertised Average Home Price by Zip code
• Filter: Zip code
• Zip code: color coded for bar graph
• Legend: Color coded by zip code
• Columns: Zip code
• Rows: Avg(Price)
• Visualization of Sheet 1:

Sheet 2

• Title: Average Home Price by Zip code (World Map)
• Filter: Zip code
• Zip code:
  • Color coded
  • Text labels per zip code
• Avg(Price): Text labels on graph
• Columns: Longitude
• Rows: Latitude
• Visualization of Sheet 2:

Sheet 3

• Title: Yearly Revenue for 2016
• Filter: Week (Date)
• Sum(price(Calendar))
  • Legend is blue color coded
  • Provides a Value Range for Revenue
  • Organized by blue color tone density
    • light tone = low value
    • dark tone = high value
• Columns: Week (Date)
• Rows: Sum(price(Calendar))
• Visualization of Sheet 3:

Sheet 4

• Title: Average Home Price by Bedroom total
• Filter: Bedrooms
• Avg(Price): labels on bar graph
• Columns: Avg(Price)
• Rows: Bedrooms
• Visualization of Sheet 4:

Sheet 5

• Title: Number of homes by Bedroom total
• Filter: Bedrooms
• CNTD(id):
  • labeled in table data visualization
  • Counts the number of homes with values that range from one to six bedrooms.
• Rows: Bedrooms
• Visualization of Sheet 5:

Completion of Tableau Dashboard:

Now that the data elements for the dashboard are organized in a concise manner, the image below represents the finalized Tableau dashboard that will be used for Airbnb home analysis that is focused on 2016 data for Seattle Residential areas.

Areas for data set improvements:

Multiple areas for data set improvements based on 2016 raw data set.

• Raw Data - Listings (worksheet)
  • Correct missing data for the following columns:
    • picture_url
    • host_response_rate
    • host_identity_verified
• Reviews (worksheet)
  • reviewer_name column
    • To obtain accurate information about reviewers, more data may need to be collected, possibly through a survey or other methods.
  • Comments column
    • Some reviewers have not provided all of the information required for their review.
      • Consumer reviews should be complete and represent accurate responses of their stay in Airbnb homes.
        • Implementation of survey enhancements is strongly recommended.
• Calendar (worksheet)
  • date column
    • January 2017 data is present in 2016 data set.
      • Talk with Airbnb clients about the situation; in which, Airbnb clients should reply on what proper course of action to take regarding the 2017 data entries.
    • Available column
      • The only values for this column are t and f
        • Talk to Airbnb stakeholders to clarify the intent of the recorded data values.
    • price column
      • There are a total of 4,417 entries that have no price data values.
        • These values also have a correlation with the Available column where data contains the value of f.
          • Talk to Airbnb stakeholders to clarify the intent of the recorded data values.

Conclusion and thankfulness for viewing data project:

Please feel free to reach out to me on LinkedIn if you have any comments or questions.

Lastly, thank you very much for viewing this data project.